The Telemark Meningococcal Project 1987-2007
Twenty years experience of preventing secondary cases of meningococcal disease by identification and eradication of the disease–causing strain of Neisseria meningitidis in close contacts of
patients with primary meningococcal disease
Master Degree Diploma in Public Health
Bjørn-Erik Kristiansen
June 2008
-Table of Contents
Acknowledgements ... 3
Introduction ... 4
Meningococcal disease... 4
Bacteriology, immunity and typing methods ... 4
Chromosomal DNA fingerprinting ... 5
PCR amplicon restriction endonuclease analyses (PCR AREA). ... 5
MLST typing. ... 6
Carriage of meningococci ... 6
Measures for preventing the spread of meningococcal disease. ... 7
Epidemiology of meningococcal disease ... 10
Meningococcal disease in the county of Telemark before the Telemark Meningococcal Project... 11
Materials and Methods ... 14
Organization of the project... 14
Interventions undertaken by the Telemark Meningococcal Project... 15
Bacteriological methods... 17
Databases and statistical calculations... 17
Results ... 20
Patients ... 20
Age. ... 20
Sex... 22
Sources of material for isolation of N. meningitidis. ... 22
Clinical outcome. ... 23
Annual number of cases. ... 24
Serogroup distribution of disease-causing strains... 25
Close contacts... 25
Bacterial findings. ... 26
In whom could the disease-causing strain be found?... 27
Odds Ratio for being a carrier of the disease-causing strain... 30
Discussion ... 31
Cost – benefit estimates and cost utility analyses ... 31
Does the Telemark Meningococcal Project prevent secondary cases? ... 37
Does the Telemark Meningococcal Project influence the number of primary cases of meningococcal cases? ... 38
Conclusions ... 44
References ... 46
Appendix A. Historical documents including updated (2003) recommendations for the Telemark Meningococcal Project... 55
Acknowledgements
This diploma was based on a project initiated by the former County Officer of Health in the County of Telemark Arne Birger Knapskog. The recommendations given by the County Health Officer were developed by a group of specialists consisting of Gunnar Hopen, consultant in infectious diseases, Telemark Central Hospital, Jon Steen Johnsen, consultant in paediatrics, Telemark Central Hospital, Tor Reiten, general practioner, Seljord and Bjørn-Erik Kristiansen, consultant in medical microbiology, Telelab AS.
At Telelab AS, my collaborators were Andrew Jenkins, Ph.D, research director, Yngvar Tveten, consultant, Linda Strand, researcher, Randi Aakre researcher, Eirik Ask, researcher and Anne Gry Allum, technician..
Thanks also to all the enthusiastic local infection control physicians and nurses who organized the collection of throat specimens from contacts as well as the information meetings. Thanks also to all the families and contacts of the patients who patiently and without exception allowed us to collect throat samples. Without their contribution there had been no “Telemark Meningococcal Project”. Finally, great thanks also go to my supervisor Bjørn Straume and the other lecturers and staff at the Institute of Community Medicine at the University of Tromsø. Their teaching and support was of great inspiration!
Introduction
Meningococcal disease
Meningococcal disease is caused by the bacterium Neisseria meningitidis and presents as septicaemia, meningitis or by a combination of septicaemia and meningitis (1, 2, 3). More seldom, meningococcal disease occurs as an entity called chronic meningococcemia characterised by fever, rash and arthritis (4). The overall fatality rate of meningococcal disease is approximately 10 %, but in septicaemia the fatality rate may reach 30 % (5, 6, 7, 8, 9, and 10).
Bacteriology, immunity and typing methods
N. meningitidis is identified by microscopy of gram stained specimens and standard biochemical tests applied on pure culture (11). The bacterium may be classified into serogroups based on antigenic differences in the capsular
polysaccharides. There are at least 10 different serogroups the most commonly occurring denoted A, B, C, Y, and W135 (11, 12). The serogroup classification is important because the capsular polysaccharides of serogroup A, C, Y and W135 are highly antigenic and exposure by invasive disease, carriage or by vaccination stimulates the formation of bactericidal antibodies which are correlated with protection against (13, 14, 15, 16, 17). There exist vaccines against these four serogroups (18, 19). The serogroup B polysaccharide is not immunogenic in humans, probably because it is expressed on brain cell in the foetus and is therefore a part of “self” (20)
Meningococci can further be divided into serotypes and sero-subtypes based on immunogenic differences in proteins of the cell wall’s outer membrane. The Norwegian group B meningococcus mainly has the serotype 15:P1.16 (21, 22).
Vaccines have been made against the serotype antigens. A nation-wide study was performed in Norway nearly 20 years ago. The vaccine was safe, but had only 57
% protection and protection lasted for only 6 months (23, 24).
Chromosomal DNA fingerprinting. The first genetic method, chromosomal DNA fingerprinting, was developed by our group at The University of Tromsø in the early eighties (13, 25, 26, 27, 28, 29). The method uses restriction
endonucleases recognising unique DNA fragments of 6 base pairs, to cleave chromosomal DNA into fragments with a mean length of approximately 4000 base pairs. The fragments are separated according to length, by gel electrophoresis followed by staining. The resulting band pattern consists of approximately 50 different bands, and each strain has its unique band pattern, comparable to the bar codes printed on almost every items that are sold today (Fig. 1). The method is laborious and takes 2 days to perform after having obtained pure culture of the bacterium.
PCR amplicon restriction endonuclease analyses (PCR AREA). PCR AREA was developed by us (21, 30, 31, 32). The method is rapid, can be
unique for each bacterial strain (Fig. 2). The PCR AREA is therefore convenient for the rapid recognition of a disease-causing strain in close contacts. By applying the same PCR method on cerebrospinal fluids from patients with meningococcal meningitis, we were able to develop the first PCR based method for diagnosis of bacterial meningitis (33).
MLST typing. However, for classification of the genus N. meningitidis, the MLST (multilocus sequence typing) has been internationally accepted as the present gold standard (34, 35, 36). Also other variants of genetic typing methods have been published (36, 37, 38)
Carriage of meningococci
N. meningitidis is part of the normal flora and most people will during their lives be a transient carrier of the bacterium (39, 40, 41). The carrier state will last for up to 1 year. During carriage the host will produce bactericidal antibodies which will protect against invasive disease (17). Carriage of meningococci may therefore be looked upon as “nature’s vaccine” against meningococci. Consequently, carriage of meningococci should never be terminated by chemoprophylaxis unless it is a disease-causing strain. These natural occurring antibodies appear around the age of 16, and increases in amount with age (13). There is an inverse relation between the protective antibodies and the age specific incidence of meningococcal disease (Fig 2). Approximately 10 % of a normal population will carry N. meningitidis at a given time. It has been shown by us (13) and in a study from the Oslo area (39), that approximately 90 % of the carrier strains are never found in patients and may be regarded as non-virulent. The remaining 10 % of the carrier strains possess a
be regarded as virulent strains. Carriers of these strains are probably the source of meningococcal disease. These carriers are at risk of developing meningococcal disease themselves or may become chronic asymptomatic carriers spreading the bacterium to susceptible persons.
Measures for preventing the spread of meningococcal disease.
Since there exist no vaccines against serogroup B meningococci, the first case of group B meningococcal disease in a community cannot be prevented. Other measures to prevent the disease from spreading must be applied. The classical way of stopping an infection by identification of the causative agent and stopping its spreading route, may be applied. Consequently, it will be necessary to identify the person(s) carrying the disease-causing strain and eradicate the disease-causing strain before it spread further. Eradication of the disease-causing strain by
chemoprophylaxis may be obtained. Chemoprophylaxis is treating a person with a short course of antibiotics to remove a potential hazardous microorganism from a a person without symptoms of disease. Rifampicin or ciprofloxacin for 2 days are most commonly used for chemoprophylaxis of meningococcal disease, and have in systematic reviews, been shown to prevent secondary cases (42, 43, 44, 45, 46, 47, 48, 49, 50). Penicillin, which is the first choice for treating patients with invasive meningococcal disease, does nor eradicate meningococci from the throat
resistance. The Norwegian group B meningococcus which is highly virulent, has developed sulphonamide resistance, probably due to overuse of this drug (6, 7, 22).
Figure 1. Fingerprint of meningococcal
chromosomal DNA
Figure 2. PCR amplicon restriction endonuclease analysis, PCRAREA
Figure 3. Plot of age specific incidence of meningococcal disease versus antibody level
The Norway recommendations for preventing spread of meningococcal disease (54) are not aimed at eradicating the disease-causing strain from the contacts, but at protect family members at the highest risk of contracting secondary infection.
The patient’s family members (household members) below 16 years of age are defined as having invasive infection regardless of clinical symptoms. They are kept at home from school or kindergarten and are treated with penicillin orally for 7 days. . However, the efficacy of this strategy is disputed as it has been
documented (51) that household members below 15 years may develop invasive disease after penicillin treatment has stopped. The reason is that the disease- causing strain has not been eradicated and therefore may infect susceptible person following the stop of penicillin treatment. The Norwegian strategy has been criticized (48, 53) for not aiming at eradication of the disease-causing strain from the environment.
Epidemiology of meningococcal disease
Meningococcal disease occurs mostly as sporadic cases in western countries, but may cause local outbreaks (2, 3, 55, 56). Endemics of meningococcal disease may occur which was the case in Norway during the period from 1974 until the late eighties (6, 56, Fig. 4). However, in the African meningococcal belt (Fig. 5), endemics are almost an annual event (2, 55, 58). The reasons may be that the mucosal surfaces of the airways are dried out making people susceptible to infections, poor health condition and low vaccine coverage. Of the 550.000 cases of meningococcal disease occurring worldwide per year, approximately 500.000
occur in Africa (57), mainly caused by the serogroup A or W135 meningococci (57, 58).
Meningococcal disease in the county of Telemark before the Telemark Meningococcal Project
The background for the Telemark Meningococcal Project was the large number of secondary cases that occurred in Telemark during the period from 1984 to 1987.
During these 4 years there were a total of 43 cases of meningococcal cases of which five were the primary cases for 12 bacteriologically verified and 4 suspected secondary cases of meningococcal disease (42). The prevalence of secondary cases of all cases was therefore nearly 30 %. At Notodden it was a large outbreak with a total of 8 cases all associated with one particular high- school. The first case appeared in March 1986 and the final in November 1986 causing widespread concern and anxiety and high consumption of antibiotics, prescribed by desperate local physicians. We had developed DNA typing methods that enabled the rapid and reliable identification of the disease - causing strain of N. meningitidis (13, 25, 26). The situation following the occurrence of a primary case of meningococcal disease is schematically visualized in Fig. 6. There will be three variants of carriers:
• Carriers of non-virulent meningococci
Figure 5. Incidences of
meningococcal disease in Africa Figure 4. Cases of meningococcal disease in Norway 1977-2008 (figures from National
Inst of Public Health, Oslo, refr.56)
Meningoccal disease Norway 1977-2008
0 50 100 150 200 250 300 350 400
1977 1979
1981 1983
1985 1987
1989 1991
1993 1995
1997 1999
2001 2003
2005 2007 Year
Cases
study we report the 20-years experiences with the Telemark Meningococcal Project and we try to evaluate:
1. if the interventions of the Telemark Meningococcal Project may have prevented secondary infection;
2. which of the contacts has the highest risk of carrying the disease-causing strain; and
3. the costs of the Project.
Figure 6. A theoretical model for the spread of the disease-causing strain at the time a case of meningococcal disease occurs in a
population (red). There are persons carrying the disease-causing strain (yellow), contacts carrying other strains of meningococci (green) as well as non-carriers (turquoise). The aim of the Meningococcal Project of Telemark is to identify the contacts carrying the disease-causing strain (yellow) and to treat these with rifampicin to eradicate it from the affected population
?
?
Materials and Methods
Organization of the project.
In 1987, the County Health Officer of the County of Telemark (165.000 inhabitants) distributed the recommendations for the Telemark Meningococcal Project to all local community infection control physicians, general practitioners, departments of medicine and paediatrics at all four hospitals of Telemark, and to the private medical microbiology laboratory of Telelab which served as the official microbiology laboratory for all hospitals, institutions and outpatients clinics in all Telemark (Appendix A). These guidelines were developed through the collaboration between consultants of the Departments of medicine and
paediatrics at the Telemark Central Hospital, the consultant at Telelab and a local community infection control physician. The recommendations contained detailed information on:
• the pre-hospital and hospital collection of specimens for bacterial diagnosis, clinical diagnosis and treatment of meningococcal disease,
• diagnostic procedures to be applied and information to be given to the local physician by the microbiology laboratory,
• procedures for identifying close contacts of patients,
• sampling of throat cultures from close contacts and registration of their relation to the patient, age and sex and telephone number in case of need for chemoprophylactic treatment
• schemes for chemoprophylactic treatment with rifampicin of contacts who carry the disease-causing strain
• post-treatment bacterial control of contacts carrying the disease-causing strain.
• information policy to the public by meetings, press-releases, and letters to family and affected community.
Interventions undertaken by the Telemark Meningococcal Project.
Typically, when a case of meningococcal disease was suspected by the consulting general practitioner, the patient was immediately admitted to hospital after blood culture and throat samples were collected. When the transport time exceeded 30 minutes, the patient was given penicillin intramuscularly or intravenously. On admission in hospital blood samples including blood culture, throat samples as well as cerebrospinal fluid were collected and sent immediately to the
laboratories. At the Telelab AS, the specimens were cultured on appropriate media and a gram stained specimen was prepared for microscopic examination. The Telemark Meningococcal Project was initiated by the microbiologist on:
• the finding of gram negative diplococci in the gram stained cerebrospinal fluid specimen (Day 0) or,
• growth of gram negative diplococci in blood culture (Day 1) or,
• growth of oxydase positive, gram negative diplococci on plates incubated
Upon confirmation of meningococci in specimen(s) from the patient, the microbiologist on duty (either of two) shall:
• inform the doctor on duty at the hospital and collect information on the patient (address, family members, whether working, in school,
kindergarten or at home, which persons have slept in the same room as the patient during the preceding 2 weeks, kissing contacts)
• alarm the local infectious disease control physician and define who are close contacts from whom throat sampling shall be performed, plan throat sampling and information meeting (Day 1 or 2)
In every case there was held an information meeting for family, contacts, and other persons affected by the outbreak with the attendance of both the
microbiologist and the local infection disease control physician. Figure 7 shows a review of how the sampling of close contacts is performed. The throat specimens were plated immediately after collection and data of each contact was collected.
Upon arrival back to the laboratory the specimens were incubated and the plates read the days thereafter (Day 2 or 3), and meningococci identified on Day 3 or 4.
DNA fingerprinting was performed on Day 3 or 4. In 1995, the method for identification disease-causing strains was changed to the PCR AREA method developed by us (30). The PCR AREA method enables the identification of the disease-causing strain to be obtained on Day 2 or 3 (Table 1). Information of whom were carriers of the disease-causing strain was given by the microbiologist to the local infection control physician, who was responsible for starting
chemoprophylactic treatment with rifampicin. Rifampicin was made available from the hospital pharmacy (Appendix A). All contacts who carried the disease-
causing strain were controlled with culture 7-10 days following chemoprophylaxis.
Bacteriological methods
Specimens from the patients (blood, cerebrospinal fluid and nasopharynx) and from the close contacts (nasopharynx) were cultured on standard media and incubated in 10% CO2 overnight as described (11). Growth of oxydase positive colonies was gram stained for gram negative diplococcic. On growth of pure culture, meningococci were identified by degradation of glucose and maltose, but not of sucrose, lactose or tributyrine. Sulphonamide susceptibility was tested by the e-test, and serogrouping was performed using antibodies against the different serogroups in a slide-agglutination test (11). DNA fingerprinting and later PCR AREA for strain identification was performed as published (25, 30). All strains of meningococci were frozen at – 700C.
Databases and statistical calculations.
The data for the patients and for the contacts were first written into Dbase III +.
For the statistical calculations the Dbase III + files were imported and converted into SPSS 15.0 for Windows (SPSS UK, Ltd., 2006). Relative risks and chi- square test were performed according to Jekel et al (64).
Day DNA fingerprinting (1987-94) PCR AREA (1995-2007)
Day 0 Identification of gram negative diplococci in CSF Identification of gram negative diplococci in CSF Information to hospital consultant Information to hospital consultant
Alarm to local infectious disease control physician Alarm to local infectious disease control physician Identification of close contacts Identification of close contacts
Planning of information meeting and throat sampling Planning of information meeting and throat sampling Day 1 Information meeting for affected population Information meeting for affected population
Throat specimen sampling of close contacts Throat specimen sampling of close contacts
Plating and incubation Plating and incubation
Day 2 Reading plates. Spreading for pure culture Reading plates. DNA extraction of oxydase positive growth Identification of disease-causing strain from PCR AREA pattern Information to carriers of disease-causing strain
Start of chemoprophylactic chemotherapy with rifampicin Day 3 Identification of meningococci
Extraction and cleavage of DNA from meningococci Start electrophoresis of DNA fragment
Day 4
Identification of disease-causing strain from DNA fingerprint pattern
Information to carriers of disease-causing strain
Start of chemoprophylactic chemotherapy with rifampicin Day
11-14 Bacterial control of contacts with disease-causing strain Bacterial control of contacts with disease-causing strain
Table 1. Steps in the Telemark Meningococcal Project from the day of microbial verification of meningococcal disease (Day 0) until bacterial control after chemoprophylaxis using either the DNA fingerprint method or the PCR AREA for strain verification
Figure 7 (row by row from top left): plates and spatula needed for collection of
nasopharynx samples, teaching close contacts how to say ”AAAAA” before collection of specimens, samples are collected using one single cotton swab which are rolled over both tonsils and nasopharynx mucosa, collection of sample by the microbiologist, incubation of plates, spreading culture for purification, growth of meningococci on chocolate plate, degradation of sugars for identification of meningococci, extraction of DNA, PCR AREA band pattern
Results
Patients
During the study period from November 1, 1987 until October 31, 2007, there were 66 cases of bacteriologically verified meningococcal disease in the county of Telemark. There was no secondary case of meningococcal disease. All 66 cases were primary cases with no link to each other.
Age. The mean age of the patients was 16,6 years ranging from 0 – 79 years (SD 22,3).
Age
79 77 69 67 66 65 54 49 39 33 20 19 18 17 16 15 11 10 8 7 6 5 4 3 2 1 0 10
8
6
4
2
0
Figure 8. Age distribution of the 66 primary cases of meningococcal disease in Telemark 1987-2007
Table. 2. Age distribution of the 66 patients
Frequency Percent Valid Percent
Cumulative Percent
0 3 4,5 4,5 4,5
1 10 15,2 15,2 19,7
2 3 4,5 4,5 24,2
3 7 10,6 10,6 34,8
4 6 9,1 9,1 43,9
5 3 4,5 4,5 48,5
6 2 3,0 3,0 51,5
7 2 3,0 3,0 54,5
8 1 1,5 1,5 56,1
10 1 1,5 1,5 57,6
11 3 4,5 4,5 62,1
15 2 3,0 3,0 65,2
16 2 3,0 3,0 68,2
17 1 1,5 1,5 69,7
18 5 7,6 7,6 77,3
19 2 3,0 3,0 80,3
20 1 1,5 1,5 81,8
33 1 1,5 1,5 83,3
39 2 3,0 3,0 86,4
49 Valid
Table 2 and Fig. 1 both show the age distribution of the 66 patients. The age-specific prevalence was highest in the youngest age group. Thirty-four (51,5 %) of the patients were six years of age or younger. There is also a high number of patients at 18 and 19 years of age (10, 6 %). Only 18,2 % of the patients were above 20 years of age.
Sex. Forty-one (62,1 %) of the 66 patients were males and 25 (37,9 %) were females (Table3).
Sources of material for isolation of N. meningitidis. Meningococci were isolated in blood culture from 48 % of the patients, from the cerebrospinal fluid in 39,4 5 of the patients (patients with meningitis), and from nasal or nasopharyngeal specimens in 12,1 % of the patients. The latter patients had all clinical signs of septicaemia (petecchial bleedings) and/or meningitis (neck-stiffness), but in whom culture from blood and cerebrospinal fluid failed due to antibiotic treatment before collecting clinical specimens.
Table 3. Sex distribution of the 66 patients
Frequency Percent Valid Percent
Cumulative Percent
Men 41 62,1 62,1 62,1
Women 25 37,9 37,9 100,0
Valid
Total 66 100,0 100,0
Clinical outcome. A total of 5 (7,6 %) of the patients died. The age of those who died (Fig. 9) was higher (mean 31,8 years, CI = 9,9 – 73,6) than in those who survived (mean 15,3 years, CI = 9,9 – 20,7). Three out of 41 (7, 3 %) males died and 2 (8 %) out of 25 females had a fatal outcome.
Table 4. Sources of material for the isolation of N.meningitidis. BK=
blood culture, HA= nasopharyngeal specimen, NE= nose, CSF=
cerebrospinal fluid
Figure 9. Age of patients who died and in patients who
survived
Age
80
60
40
20
65 5018
13 26
61 8
Frequency Percent Valid Percent
Cumulative Percent
BK 32 48,5 48,5 48,5
HA 7 10,6 10,6 59,1
NE 1 1,5 1,5 60,6
SP 26 39,4 39,4 100,0
Valid
Annual number of cases. The annual numbers of bacteriologically verified cases are shown in Table 5. The annual numbers of notified cases to the
Norwegian notification system for infectious diseases (MSIS, National Institute of Public Health, Oslo) are shown in the same table. The discrepancy between the annual figures may be due to the fact that both suspected and verified cases are notifiable to the National Institute of Public Health.
Table 5. Annual number of
bacteriological verified (Telelab AS) and notified cases (National Inst Public
Health, Oslo) of systemic meningococcal disease in the County of Telemark 1987 - 2007
Year Verified Notified
1987* 3 3
1988 8 10
1989 2 2
1990 6 7
1991 7 8
1992 7 9
1993 2 2
1994 3 2
1995 3 4
1996 7 8
1997 4 4
1998 3 3
1999 2 2
2000 3 3
2001 2 2
2002 2 2
2003 1 1
2004 0 0
2005 1 1
2006 0 0
2007 0 0
Total 66 73
Serogroup distribution of disease-causing strains. Table 6 shows that the predominant serogroup among the 66 case strains was the serogroup B causing disease in almost 65 % of the patients. Serogroup C was isolated from
approximately 25 % of the patients.
Serogroups
Close contacts
A total of 2252 close contacts of the 66 patients were identified and included. In two (3 %) out of the 66 patients, no close contacts were identified. The disease- causing strain was isolated from close contacts of 37 (56, 1 %) out of the 66
Group Percent Valid Percent
Cumulative Percent
A 1 1,5 1,5 1,5
B 42 63,6 63,6 65,2
C 17 25,8 25,8 90,9
W135 2 3,0 3,0 93,9
Y 4 6,1 6,1 100,0
Valid
Table 6. Distribution of serogroups among the disease-causing meningococcal strains from the 66 patients
Bacterial findings. The number of close contacts carrying a strain of
N.meningitidis was 302 (13, 4 %). Other oxydase positive diplococcal species (Moraxella catarrhalis and Neisseria lactamica) were found in 243 (10, 8 %) contacts (Table 7). Of the 302 isolates of N.meningitidis, 70 (23,2 %) were identical to the disease-causing strain (Table 8). The carriage rate of the disease-
causing strain among all close contacts was 3, 1 % (70/2252). The carriage rate of non-disease causing meningococci among all close contacts was 10, 3 %. Of the 70 close contacts who carried a disease-causing strain, 26 were females (37, 1 %) and 44 (62,9 %) were males. Among the 232 who carried a non-disease causing meningococcal strain, 103 (44,4 %) were females and 129 (55,6 %) were males.
Table 7. Bacterial species isolated from the nasopharynx of 2252 close contacts. M.CA= Moraxella catarrhalis, N.LA= Neisseria lactamica, N.MC= Neisseria meningitidis,
Frequency Percent Valid Percent
Cumulative Percent
1707 75,8 75,8 75,8
M.CA 79 3,5 3,5 79,3
N.LA 164 7,3 7,3 86,6
N.MC 302 13,4 13,4 100,0
Neg
Total 2252 100,0 100,0
In whom could the disease-causing strain be found?
The close contacts were divided into 20 contact groups according to their relation to the patient (Table 8). Groups 1-17 consists of persons who had direct contacts with the patient (primary contacts). Whenever a primary contact was shown to carry a disease-causing strain, his or hers kissing contacts and household-members were also screened for meningococci colonization. These contacts were defined as secondary contacts and were categorized into contact group 18. Tertiary contacts (contact group 19) were household-members and kissing contacts of secondary contacts that carried the disease-causing strain and finally, quartary contacts (contact group 20) were similarly related to tertiary contacts. The 6 first groups of contacts are household - members or kissing contact. In these 6 groups, containing 210 persons, the prevalence of the disease-causing strain was 13,3 % whereas the prevalence of non-disease causing meningococci was 7,6 %, giving a total prevalence of meningococcal carriage of 20,9 %. Among the remaining 1917 primary contacts the carrier rate of the disease-causing strain in the nasopharynx was 1, 8 % and of non-disease-causing isolates 10, 7 %.
Figure 11. A summary of the finding of disease-causing strain in primary, secondary, tertiary and quartary contacts of patients with meningococcal disease
Far Mor
Kusine Bror
12.9
14.9
15.9 11.9
Figure 10. The result of the environmental study in a kindergarten following a case of meningococcal septicemia in a 2 years old girl (red circle) diagnosed on Sept. 11 (!). On Day 1 (Sept. 12), we identified two other children in the kindergarten carrying the same disease-causing strain (red triangles) as well as one adult employee (red squares). Further investigations of these three contacts revealed that the mother and grandmother of one child and the teenage son of the adult employee harbored the disease-causing strain. By extended studies of these person’s household-members and kissing contacts, no more carriers of the disease-causing strain was revealed. All carriers of the disease-causing strain were given rifampicin with good effect
Household-kissing contacts
Pat.
Other close contacts 1,8 % (34/1917)
13,3 % (28/210) 6,6 % (8/121)
2º, 3º and 4º contacts
Contact
group Relation to the patient Frequency Percent Identical
Identical
% Different Different %
1 Fathers 49 2,2 6 12,2 3 6,1
2 Mothers 51 2,3 6 11,8 3 5,9
3 Sisters 37 1,6 3 8,1 1 2,7
4 Brothers 39 1,7 7 17,9 3 7,7
5 Kissing contacts 10 0,4 4 40,0 0 0,0
6 Others 24 1,1 2 8,3 6 25,0
Sum household members 210 9,3 28 13,3 16 7,6
7 Grandparents 56 2,5 4 7,1 6 10,7
8 Playmates family 116 5,2 2 1,7 20 17,2
9 Playmates 286 12,7 6 2,1 48 16,8
10 Nursery employees 114 5,1 2 1,8 4 3,5
11 Childminders 6 0,3 1 16,7 1 16,7
12 Other family 214 9,5 5 2,3 37 17,3
13 Classmates 342 15,2 5 1,5 48 14,0
14 Children at nursery 421 18,7 2 0,5 11 2,6
15 Teachers 65 2,9 0 0,0 3 4,6
16 Colleagues 3 0,1 0 0,0 2 66,7
17 Others 294 13,1 7 2,4 25 8,5
Sum other primary
contacts 1917 85,1 34 1,8 205 10,7
Total close contacts 2127 94,4 62 2,9 221 10,4
18 Secondary contacts 86 3,8 5 5,8 9 10,5
19 Tertiary contacts 16 0,7 3 18,8 2 12,5
20 Quartary contacts 19 0,8 0 0,0 0 0,0
Sum 2248 99,8 70 3,1 232 10,3
Missing 4 0,2
Total 2252 100 70 3,1 232 10,3
Table 8. The number of contacts in different groups according to relation to the patient.
“Identical” denotes a contact carrying a disease causing strain of N. meningitidis, and “Different”
denotes carriage of a strain that is different from the disease-causing strain.
In the secondary carriers, the carriage rate of the disease-causing strain was 5, 8 %, among tertiary contacts 18, 8 % and in quartary contacts 0 %.
Odds Ratio for being a carrier of the disease-causing strain. From Table 8 it can be calculated that the Odds Ratio (OR) for carrying the disease-causing strain being a house- member or a kissing contact (contact groups 1-6) is 8, 52. These contacts, therefore, have 8, 52 times higher risk than other primary contacts, to carry the disease-causing strain. The chi- square value was calculated to 15, 31 which gives shows that the difference in Odds Ratio is statistical significant at p < 0, 0005 level with one degree of freedom (df=1). Furthermore, the attributable risk percentage in the exposed (AR %) can be calculated from the formula:
AR % = Risk (exposed) – Risk (unexposed)/Risk (exposed) X 100
where “exposed” are the household-members and kissing contacts, and “non-exposed” are other primary contacts. The AR % is (13,3 – 1,8)/13,3 X 100 = 86,5 %, which means that of the total risk of carrying the disease-causing strain, being a household-member or a kissing contact, 86,5 % of the risk stems from being in contact groups 1-6, the remaining risk being caused by other factors.
Discussion
Cost – benefit estimates and cost utility analyses
The costs of the Project are mainly linked to the laboratory analyses. The flow- chart for the procedures leading from sampling and culture to conclusive
“identity” or “not-identity” for the nasopharyngeal specimens from all 2252 contacts, is shown in Fig 12. Each procedure has its own price determined by the Ministry of Health. Hence, it is possible to estimate the total costs of the Project.
All prices and costs are given in 2007 - values. The total costs of the project are estimated to 660.000 NOK (Table 9A). What has been gained by this money? A saving of the Project is the hospitalization cost of the persons who were prevented from contracting meningococcal disease. How many cases were prevented by our interventions? The number of cases of meningococcal that may have been
prevented by the Project can be calculated from the expected prevalence of secondary cases. Given that 66 is the number of primary cases, the expected total number of cases is 66 plus the expected number of secondary cases. In the
literature, the prevalence of secondary infections is reported to vary between <2 % (65) and 30 % (42) with 10 % reported from all Norway (66). Given a prevalence of secondary cases at 2 %, the expected number of cases is 67,3; hence 1
secondary infection may have been prevented. Given a prevalence of secondary
presently at 33647 NOK. Each diagnosis is given a Diagnosis Related Group (DRG) point according to estimated costs for that particular diagnosis (laboratory tests, X-ray procedures, treatment procedures and drugs). According to this system, meningococcal disease has three different cost categories or DRG points:
• Infections in the central nervous system: 2, 21 points.
• Septicemia below 17 years of age: 2,44 points
• Septicemia in patients 17 years of age or older: 1,94
The savings of hospitalization costs varied from approximately 82.000 NOK (1 secondary case prevented) to approximately 470.000 NOK (7 patients prevented, Table 9B). It was assumed that of the 7 patients that may have been prevented, 3 had septicemia and were below 17 years, 3 had meningitis, and one patient was above 17 years and had septicemia. The saving in hospitalization costs, therefore was less than the costs of the Project (660.000 NOK).
But in addition to saving costs for hospitalization, which can be said to be the cost-benefit effect of the Project, the patients who may have been prevented from contracting disease, may also benefit by not having its health related life quality reduced following meningococcal disease. In the worst case meningococcal disease may be fatal, and 30 % of those with septicemia die (10), the overall mortality is 10 %. Consequently, there may be a benefit from the Project that cannot be measured in money, prevention of reduced health quality of life. This may be called the cost – utility effect and can be measured by cost – utility analyses. We apply the QALY (quality adjusted life years) principles to calculate
the improvement in health related quality of life that is assumed to be gained from our interventions. Our calculations are based on some assumptions:
• The life-expectancy of patients who survive meningococcal disease is not shortened
• There is a small, but significant risk for patients who recover from meningococcal disease to get permanent sequelae (68, 69, 70), the most common being psycho-social, physical and orthopedic problems and some few cases with epilepsy, hearing loss and blindness.
• Patients who contract meningococcal disease had no reduction in health related quality of life before contracting meningococcal disease
• All patients who may have been prevented survived with no fatal case.
Based on the literature reporting sequelae, we estimate that the reduction in health associated QALY in patients recovering from meningococcal disease to be 0, 2 on a range from 0 - 1 where 1 represents no reduction in health-related life quality and = is death. Given that the age of the prevented cases corresponds to the mean age of the 66 patients (17 years), their remaining life span can be calculated by means of the calculator that is found on the homepage of Statistisk Sentralbyrå (http://www.ssb.no/vis/emner/02/02/10/dode/art-2008-04-10-01.html). If the patients is a woman at 17 years of age, she will have a life-expectancy of 82 years (and 2 months) living in Telemark. Being a male at 17 years of age, the life-
([[TF * (H1 – H0)] * NF] + [TM * (H1 - H0)] * NM]])
2 %prevalence([[TF * (H1 – H0)] * NF] + [TM * (H1 - H0)] * NM]])
10 % prevalencewhere H1 is the health quality without contracting secondary disease which is assumed to have the value “1”, H0 is the quality of life after recovering from secondary disease and is set to “0, 8”, TF ( 65 years) is the remaining life expectancy for females at 17 years of age, and TM (60 years) for males at 17 years of age, living in Telemark, NF is the number of female patients who were prevented, and NM the number of males prevented from secondary infection.
(65 * 0,2 * 0 + 60 * 0,2 * 1)
2 % level= 0 + 12 = 12 QALYS (65 * 0,2 * 3 + 60 * 0,2 * 4)
10 % level= 29 + 48 = 77 QALYS
Given a prevalence of secondary infections between 2 % and 10 %, the health quality gain from the Project are in the magnitude of from 12 – 77 QALYS.
The Telemark Meningococcal Project may have additional positive effects that are difficult to measure. The affected population is offered immediate oral and written information on meningococcal disease which may lead to quicker response once a secondary case occurs. Moreover, information and the fact that something active is done (bacterial testing), may prevent much of the anxiety that always follows the footsteps of meningococcal disease. The local health system which is under significant pressure in this case, gets active support from the Specialist Health Care System. Local physicians have expressed a relief in their situation as a
consequence of the Project. There are less telephones and visits than before the Project.
Culture of nasopharyngeal specimens on CV plates 2252
No growth 1707 Growth. Oxydase testing
545
PCR AREA of suspected growth
152
Sugar degradation 545
Neisseria meningitidis 302
Serogrouping 302
Sulphonamide susceptibility 302
DNA fingerprinting 302
Procedure
Price (NOK)
No of tests
Total (NOK)
Culture of meningococci 83 2252 186916
Oxydase testing 38 545 20710
Gram staining 38 545 20710
Sugar fermentation 38 545 20710
Sulphonamide susceptibility testing 38 302 11476
Agglutination (3 serogroups) 114 302 34428
DNA fingerprinting, alone (until 1994) 909 150 136350
DNA fingerprinting in combination with PCR AREA (from
1995) 909 152 138168
PCR AREA (from 1995) 606 152 92112
Total costs of the Telemark Meningococcal Project 661580 Table 9A. Estimates of the costs of the Project. The costs are mainly linked to the laboratory procedures.
At 2 % prevalence level At 10 % prevalence level Diagnosis Related Group Unit price
DRG
points No. patients Costs (NOK) No. patients Costs (NOK)
20. Infection of CNS 33647 2,21 0 0 3 223079,61
416. Septicemia > 17 years of age 33647 1,94 0 0 1 0
417. Septicemia 17 years or younger 33647 2,44 1 82098,68 3 246296,04
Total cost of hospitalization of prevented cases 82098,68 469375,65
Table 9B. Estimates of the costs of hospitalization of patients who were prevented from contracting secondary meningococcal disease by the intervention measures of the Project. At a level of 2 % prevalence of secondary infection, the estimated number of prevented cases is 1, and at the level of 10 %, 7 cases are prevented (see Discussion)
Does the Telemark Meningococcal Project prevent secondary cases?
During the Project period, there was no secondary case of meningococcal disease, but 66 bacteriological verified primary cases. A vital question is if this absence of secondary cases was a consequence of our intervention, or whether it can be explained by chance. As was discussed in the section above, the expected total number of cases of meningococcal disease is 66 X 100/98 = 67, 3 cases at a 2 % level of secondary cases. At the 10 % level, the expected number of total cases is 66 X 100/90 = 73, 3 cases. Consequently, the number of cases which could have been prevented by our interventions varies between 1, 3 and 7, 3 cases. At the 30
% level, the number of prevented cases would have been 28, 3. We find it not unlikely that these expected secondary cases have been prevented by our interventions. By applying the chi-square test on the differences between the number of cases that were identified (66 cases), and the expected number of cases, calculations show that the numbers are too small to demonstrate any difference of statistical significance at the 2 % or 10 % level, respectively. Only at a level of 15
% prevalence of secondary cases, would the material allow the difference to be statistical significant. At the earlier reported level of 30 % prevalence from Telemark (42), the chi-square test shows a statistical significant difference at p =
Does the Telemark Meningococcal Project influence the number of primary cases of meningococcal cases?
Previous studies from Norway show that only a few different virulent strains are causing meningococcal disease (21, 30). These clones probably circulate in the population constantly. By our interventions, the prevalence of these virulent clones may be reduced to such an extent that infecting chains are broken resulting in a reduction also of primary cases. The encounter between a virulent
meningococcus and a susceptible host occurs less frequent because the virulent clone is “diluted” in the population; the “dilution” effect. This “dilution” effect is difficult to prove since it is known that there has been a decline in the incidence of meningococcal disease in all Norway during the last 20 years, probably caused by a rise in herd immunity against the circulating clones.
To try to answer the question whether the interventions of the Telemark
Meningococcal Project also reduced the number of primary cases, we compared the development of incidence of meningococcal disease in Norway, Telemark and the neighbouring county of Vestfold (220.000 inhabitants) during three 10-years periods:
1. 1978-1987; the 10-years period immediately prior to the initiation of the Project
2. 1988-1997; the first 10-years period after the Project was started 3. 1998-2007; the second 10-years period after initiation of the Project While the effect on secondary cases is assumed to have immediate effect on the occurrence of meningococcal disease, the “dilution” effect will probably appear after several years of eradicative interventions. The reason for the late effect is
that it takes time to eradicate the virulent clones enough to block the routes of spread. We hypothesize that a reduction in the mean annual incidence from the first 10-years period to the second 10-years period after the project was started, has two explanations:
• An increase in herd immunity leading to a general decline affecting also neighbouring counties
• The “dilution” effect caused by systematic and thorough eradication of virulent clones of the Project
The “dilution” effect can only be seen in Telemark, and the decrease in
meningococcal disease occurrence shall be more pronounced in Telemark than in neighbouring counties. We therefore compared the incidences of meningococcal disease in Telemark with that of neighbouring counties of Vestfold, Buskerud and Aust-Agder (Table 10 and Fig. 13) in the two 10-years periods after the Project was started. Fig. 14 shows a map of the region. The incidence of meningococcal disease in Telemark is constantly lower than in all Norway and in Vestfold during the last 10-years period of the Project (1998-2007). As can be seen from Fig. 15 and Table 11 , the mean annual incidence in Telemark in the first 10-years period of the Project (1988-1997) was 3, 52 cases/100.000 inhabitants (CI 2, 13 – 4, 91), and in the second 10-years period the mean annual incidence was 0, 86 (CI 0, 34 – 1, 38). Since the confidence intervals do not overlap, the fall in the mean
significant fall in meningococcal disease incidence seen in Telemark from the period 1987-1996 to 1997 – 2007.
Table 10. Incidence of meningococcal disease in all Norway, in Telemark and in each of its neighbouring counties Buskerud, Vestfold and Aust.Agder 1978 – 2007 as well in these neighbouring counties together (BVAA). The cases are notified cases by the National Institute of Public Health, Oslo. No: number of cases, Popul:
population in 100.000, Incid: incidence (No cases/100.000 inhabitants/year)
County 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
All No. 270 323 230 257 267 368 302 317 263 259 177 176 165 167 201 126 102 158 139 115 105 83 83 78 51 48 37 39 35 30
Norway Popul 40,8 40,8 40,9 40,9 41,1 41,3 41,5 41,6 41,8 41,9 42 42,2 42,4 42,5 42,7 43 43,2 43,5 43,7 43,9 44,2 44,5 44,8 45 45,2 45,5 45,8 46,1 46,4 46,8
Incid 6,6 7,9 5,6 6,3 6,5 8,9 7,3 7,6 6,3 6,2 4,2 4,2 3,9 3,9 4,7 2,9 2,4 3,6 3,2 2,6 2,4 1,9 1,9 1,7 1,1 1,1 0,8 0,8 0,8 0,6
Buskerud No. 11 11 10 12 17 5 11 14 13 21 6 9 8 8 8 8 9 7 5 4 5 4 5 5 3 1 3 2 1 1
Popul 2,1 2,1 2,1 2,2 2,2 2,2 2,2 2,2 2,2 2,2 2,2 2,2 2,2 2,3 2,3 2,3 2,3 2,3 2,3 2,3 2,3 2,3 2,4 2,4 2,4 2,4 2,4 2,4 2,5 2,5
Incid 5,2 5,2 4,8 5,5 7,7 2,3 5 6,4 5,9 9,5 2,7 4,1 3,6 3,6 3,5 3,5 3,9 3 2,2 1,7 2,2 1,7 2,1 2,1 1,3 0,4 1,3 0,8 0,4 0,4
Vestfold No. 15 19 19 18 11 16 18 21 20 14 11 5 5 8 6 5 3 1 9 5 5 7 2 5 4 2 2 4 3 3
Popul 1,8 1,9 1,9 1,9 1,9 1,9 1,9 1,9 1,9 1,9 1,9 1,9 2 2 2 2 2 2 2 2,1 2,1 2,1 2,1 2,2 2,2 2,2 2,2 2,2 2,2 2,2
Incid 8,3 10 10 9,5 5,8 8,4 9,5 11,1 10,5 7,4 5,8 2,6 2,5 4 3 2,5 1,5 0,5 4,5 2,4 2,4 3,3 1 2,3 1,8 0,9 0,9 1,8 1,4 1,4
Aust- No. 3 1 1 4 2 3 1 3 2 1 4 5 3 5 7 4 4 6 7 0 3 0 5 2 1 2 0 2 1 3
Agder Popul 0,9 0,9 0,9 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Incid 3,3 1,1 1,1 4 2 3 1 3 2 1 4 5 3 5 7 4 4 6 7 0 3 0 5 2 1 2 0 2 1 1
BVAA No. 29 31 30 34 30 24 30 38 35 36 21 19 16 21 21 17 16 14 21 9 13 11 12 12 8 5 5 8 5 7
Popul 4,8 4,9 4,9 5,1 5,1 5,1 5,1 5,1 5,1 5,1 5,1 5,1 5,2 5,3 5,3 5,3 5,3 5,3 5,3 5,4 5,4 5,4 5,5 5,6 5,6 5,6 5,6 5,6 5,7 5,7
Incid 6 6,3 6,1 6,7 5,9 4,7 5,9 7,5 6,9 7,1 4,1 3,7 3,1 4 4 3,2 3 2,6 4 1,7 2,4 2 2,2 2,1 1,4 0,9 0,9 1,4 0,9 1,2
Incidence meningococcal disease Norway, Vestfold and Telemark 1978 - 2007
0 2 4 6 8 10 12
1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 Year
Incidence (cases/100.000 inhab/year)
Norway Telemark Vestfold
Figure 13. Incidence of meningococcal disease in all Norway, and in the counties of Vestfold and Telemark 1978-2007
Figure 14. The county of Telemark and its surrounding counties. The closest contact is between Telemark (165.000 inhabitants) and Vestfold (220.000 inhabitants).
1978-1987 1988-1997 1998-2007 All Norway
Confidence level (95 %) 0,71 0,54 0,44
Upper confidence interval
level 7,63 4,1 1,75
Mean 6,92 3,56 1,31
Lower confidence interval
level 6,21 3,02 0,87
Telemark
Confidence level (95 %) 2,2 1,39 0,52
Upper confidence interval
level 7,53 4,91 1,38
Mean 5,33 3,52 0,86
Lower confidence interval
level 3,13 2,13 0,34
Vestfold
Confidence level (95 %) 1,14 1,08 0,55
Upper confidence interval
level 10,19 4,01 2,27
Mean 9,05 2,93 1,72
Lower confidence interval
level 7,91 1,85 1,17
7,63 7,53
10,19 9,05 8 7,91
10 12
0 inhab/year)
Norway Telemark Vestfold Figure 15. Mean annual incidence and confidence intervals (CI) of
meningococcal disease in three 10-yeras periods: Per 1: 1978-1987, Per 2:
1988-1997, Per 3: 1998-2007. The figures are notified cases to the National Inst Publ Health, Oslo
Table 11. Annual mean incidences with confidence intervals for meningococcal disease during three 10-years periods in Norway, Vestfold and Telemark 1978-2007
Conclusions
• There has been no secondary case of meningococcal disease in the county of Telemark after the Telemark Meningococcal Project was started in November 1987.
• It is not unlikely that the Project has prevented from 1 to 7 cases of secondary meningococcal infections.
• The prevalence of secondary cases was reduced from 30 % prior to the project (1984-87) to 0 % after the Project was started.
• The annual incidence of meningococcal disease in Telemark has been constantly lower in Telemark during the last 10 years compared to the incidence in all Norway and in Vestfold.
• There was a statistical significant fall in the mean annual incidence of notified meningococcal disease from the first 10-years period of the
project (1988-97) to the second 10-years period (1998-2007). No statistical significant fall was seen in the neighboring counties of Vestfold, Buskerud or Aust-Agder.
• The risk of being a carrier of the disease-causing strain is highest among house-hold members and kissing contacts. The administration of
chemoprophylaxis to these contacts once meningococcal disease occurs should be implemented in Norwegian recommendations.
• The costs of the Project are higher than what is saved hospitalization expenses for cases that may have been prevented.
• The Project may have saved from 12 – 99 quality adjusted life years (QALY)
• The Project should be continued and its recommendations implemented on a permanent basis for preventing meningococcal disease to spread in Telemark
